Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

[Phys-L] Re: Reaction Time (was Re: Human Error?)



I'll have to admit to having spaced out in the middle of this thread, but
the post from George and John below suggests a reply (that may be repeating
what others have said).

If 'reaction time' is a factor in an experiment, then:

a) It is uncertainty in timing due to reaction that is a factor--not 'human
error'.
b) As John suggests, one then needs to look at the experiment more closely
to see if that uncertainty can be eliminated or minimized.

Towards the latter, the PENDULUM experiment is almost classic. Once we
determine what factors (length, mass, amplitude of swing) might effect the
motion, it then becomes the problem of just how are we going to time the
swing. I'll setup a reasonably long pendulum, get it swinging and then with
a stopwatch, I'll time one swing. I'll then ask the students if they are
satisfied with that. A few are, but most say no--we need to repeat that
measurement a bunch of times and average. OK, I'll say, but WHY are you
unhappy with my single measurement? Someone will always offer--well you
might not have started and stopped the stopwatch precisely when the swing
was at the maximum of the swing. Ok--does repeating this process five times
help? Yes is the first answer. Then I say, what if I always start the
watch too soon and stop too late? Most are stumped then. But with a little
more prodding to suggest a better technique, you can get someone to suggest
timing a sequence of swings. Then ask--why is that better? Ultimately then
you get at the answer that you introduce one uncertainty but average it over
x-swings. By making x large you can reduce the uncertainty _almost_ to
zero.

Apply the same reasoning to the falling ball, and you come up with the need
for more precise timing techniques. This is when we can pull out our Pasco
time of flight equipment and turn to electronic timing--both release and
arrival times. You can also look toward timing motion on the incline (say a
tilted air track--a la Galileo) to slow the motion so as to reduce the
percentage uncertainty in the timing--especially if it must be done by hand.

So, the 'human factor' may introduce UNCERTAINTIES in the data (not errors),
but proper experimental design should be able to reduce those uncertainties
to the order of the statistical uncertainties of the equipment.

Rick

*********************************************************
Richard W. Tarara
Professor of Physics
Saint Mary's College
Notre Dame, Indiana
rtarara@saintmarys.edu
********************************************************
Free Physics Educational Software (Win & Mac)
NEW: Animations for PowerPoint
Newton's Maze lab simulation
Simplified Energy Management Simulator
www.saintmarys.edu/~rtarara/software.html
Energy 2100--class project
www.saintmarys.edu/~rtarara/ENERGY_PROJECT/ENERGY2100.htm
********************************************************
----- Original Message -----
From: "Spagna Jr., George" <gspagna@RMC.EDU>
To: <PHYS-L@LISTS.NAU.EDU>
Sent: Monday, May 02, 2005 7:04 AM
Subject: Re: Reaction Time (was Re: Human Error?)


John wrote:

-----Original Message-----
From: John Mallinckrodt [mailto:ajm@CSUPOMONA.EDU]
Sent: Friday, April 29, 2005 8:40 PM
To: PHYS-L@LISTS.NAU.EDU
Subject: Reaction Time (was Re: Human Error?)


If reaction time is genuinely a source of error in such experiments
then you need to go back to the drawing board and redesign the
experiment. Reaction time can only affect such measurements if, for
instance, a) you let your partner drop the ball while you run the
stopwatch AND you insist that your partner give you no warning or b)
you consciously refuse to watch the ball as it drops and wait for the
sound of the ball hitting the floor before stopping the watch.

Of course there will be variations in the times obtained from
successive time trials, but I maintain that they have essentially
nothing to do with the phenomenon that is properly called "reaction
time."

Typical reaction times are on the order of .25 s; the stopwatches we use
display precision on the order of +/- 5 ms. I suggest to my students
that claiming the instrumental limit as the uncertainty of their
measurement is unreasonable, especially for a single trial. Repeated
trials characteristically show a standard deviation consistent with
reaction time, rather than the stopwatch's displayed precision.

********************************************
"The wise person doesn't give the right
answers, but poses the right questions."
- Claude Levi-Strauss
********************************************

Dr. George Spagna
Chair, Physics Department
Randolph-Macon College
P.O. Box 5005
Ashland, VA 23005-5505

phone: (804) 752-7344
fax: (804) 752-4724
e-mail: gspagna@rmc.edu
http://faculty.rmc.edu/gspagna
_______________________________________________
Phys-L mailing list
Phys-L@electron.physics.buffalo.edu
https://www.physics.buffalo.edu/mailman/listinfo/phys-l